Skip to Main Content
Field workforces have to implement day-to-day control decisions to keep the large scale real-world systems (e. g. Water Distribution Systems) operational. However, existing customized Decision Support Tools (DSTs) for field workforces operate on static and predefined datasets, available in the workers' devices. This results in planning control actions that may appear safe in the local context but can cause conflicts and disturbances elsewhere in the system. To enable informed decision making, we discuss a generic architecture that supports the development of dynamic predictive DSTs for the field workforces. The predictive DSTs would not only provide up-to-date information but also allow field workforces to perform `What-If' simulations. We extend our previous work, on accessing simulations via lightweight mobile devices, by updating the `What-If' simulations with the real-time Wireless Sensor Network (WSN) data and the scheduled future control actions. The calibration with the sensor network data enables the simulations to predict the real-time state of the system, while the incorporation of future control actions reduces the risk of unexpected feedback between different parts of the system. Hence, considering both the current and possible future state in simulations enables the field workforces to take informed control decisions in the real-world system.